I'm trying to find limit distribution of $\sqrt{n}(Y_n-e)$ as $n \to\infty$, where:

$$Y_n=\left(\prod_{i=1}^n U_i\right)^{-1/n}$$ and $U_i$ is i.i.d. uniform distributions in interval $(0,1)$.

My thinking so far is to use formula for uniform distribution product:

$$P_{U_1\cdots U_n}(u)=\frac{(-1)^{n-1}}{(n-1)!}(\ln u)^{n-1}$$

and then by using limit rules get the final solution.

However, I feel that the path I have chosen is very complicated and I wonder is there a more elegant solution?


The central limit theorem will give you the limit distribution for $\sqrt n ( \log Y_n - d)$ for some $d$ that depends on the distribution of $\log U_n$. Then apply the ``delta method'' (to the function $x\mapsto \exp(x)$) to finish the job.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.